EP0251278A2 - Bildverarbeitungsverfahren und Vorrichtung dazu - Google Patents

Bildverarbeitungsverfahren und Vorrichtung dazu Download PDF

Info

Publication number
EP0251278A2
EP0251278A2 EP19870109332 EP87109332A EP0251278A2 EP 0251278 A2 EP0251278 A2 EP 0251278A2 EP 19870109332 EP19870109332 EP 19870109332 EP 87109332 A EP87109332 A EP 87109332A EP 0251278 A2 EP0251278 A2 EP 0251278A2
Authority
EP
European Patent Office
Prior art keywords
color
image
data
density
image processing
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
EP19870109332
Other languages
English (en)
French (fr)
Other versions
EP0251278B1 (de
EP0251278A3 (en
Inventor
Masahiko Konishiroku Photo Ind. Co. Ltd Matsunawa
Hiroyuki Konishiroku Photo Ind. Co. Ltd Yamamoto
Hiroshi Konishiroku Photo Ind. Co. Ltd Niki
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Konica Minolta Inc
Original Assignee
Konica Minolta Inc
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority claimed from JP61154668A external-priority patent/JPS6310884A/ja
Priority claimed from JP61154672A external-priority patent/JPH0628378B2/ja
Priority claimed from JP61154669A external-priority patent/JPS6310885A/ja
Priority claimed from JP61154667A external-priority patent/JPS6310883A/ja
Priority claimed from JP61154670A external-priority patent/JPS6310886A/ja
Priority claimed from JP61154666A external-priority patent/JPS6310882A/ja
Application filed by Konica Minolta Inc filed Critical Konica Minolta Inc
Publication of EP0251278A2 publication Critical patent/EP0251278A2/de
Publication of EP0251278A3 publication Critical patent/EP0251278A3/en
Application granted granted Critical
Publication of EP0251278B1 publication Critical patent/EP0251278B1/de
Anticipated expiration legal-status Critical
Expired - Lifetime legal-status Critical Current

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N1/00—Scanning, transmission or reproduction of documents or the like, e.g. facsimile transmission; Details thereof
    • H04N1/46—Colour picture communication systems
    • H04N1/64—Systems for the transmission or the storage of the colour picture signal; Details therefor, e.g. coding or decoding means therefor
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N1/00—Scanning, transmission or reproduction of documents or the like, e.g. facsimile transmission; Details thereof
    • H04N1/46—Colour picture communication systems
    • H04N1/56—Processing of colour picture signals
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N1/00—Scanning, transmission or reproduction of documents or the like, e.g. facsimile transmission; Details thereof
    • H04N1/46—Colour picture communication systems
    • H04N1/56—Processing of colour picture signals
    • H04N1/58—Edge or detail enhancement; Noise or error suppression, e.g. colour misregistration correction

Definitions

  • the present invention relates to an image proces­sing method for processing a color image and an apparatus therefor.
  • An image processing apparatus is an apparatus for reading out an original image in a picture element (or pixel) unit with a photo-electric transducer such as a CCD and subjecting it to an image processing by coding it in a multiple-valued notation (including the binary notation, as follows) or the like.
  • the optical image from an original image is separated into a plurality of colors (e.g., Red and Cyan) with a color-­separating optical system and is subjected to the image processing specified above.
  • Fig. l presents color charts depicting the spectral reflectivity characteristics of the achromatic, blue and red colors, respectively.
  • Figs. l(A), l(B) and l(C) show the respective reflectivity characteristics (%) of the achromatic, blue and red colors against the wavelength (nm).
  • curve fl plots the charac­teristics of a sample whereas curve f2 plots the charac­teristics of a reference density plate (or a white density plate).
  • luminance signal information can be expressed as follows: V R + V C (2).
  • V R ⁇ /V RW ⁇ V C ⁇ /V CW ⁇ V R ⁇ V C Since a chromatic image exhibits peculiar character­istics: for red colors: V R ⁇ /V RW > V C ⁇ /V CW ⁇ V R > V C ; and for cyan colors: V R ⁇ /V RW ⁇ V C ⁇ /V CW ⁇ V R ⁇ V C (4).
  • the ratio V R /V C or V C /V R can be simply adopted as the axis expressing the difference in colors.
  • the color extraction can be made as follows: for black colors: V R /V C ⁇ l.0, or V C /V R ⁇ l.0; for cyan colors: 0 ⁇ V R /V C ⁇ l.0, or l.0 ⁇ V C /V R ⁇ ⁇ ; and for red colors: l.0 ⁇ V R /V C ⁇ ⁇ , or 0 ⁇ V C /V R ⁇ l.0 (5).
  • the handling is inconvenient because a symmetry is not obtained with respect to an achromatic axis Z.
  • Fig. 2 the handling is inconvenient because a symmetry is not obtained with respect to an achromatic axis Z.
  • the ordi­nate indicates the luminance information (V R + V C )
  • the abscissa indicates the color difference information (V R /V C ).
  • the color difference information (V R /V C ) diverges to infinity, as is apparent from the formulas (5). It is, however, understood that the black, cyan and red colors can be extracted by thus determining any ratio between the image data V R and V C . Therefore, the color extractions by the differences in the ratio of the red and cyan components contained in the whole luminance signal will be examined.
  • V R ⁇ V C holds, as expressed in the foregoing equation (3). Therefore, the quantities of the image data V R and V C contained in the luminance signal, i.e., the total quantity of light (V R + V C ) are equal. In other words, the following equations hold in the black (or achromatic) colors: V R /(V R + V C ) ⁇ 0.5; and V C /(V R + V C ) ⁇ 0.5 (6). In a chromatic color, on the contrary, the quantities of the image data V R and V C contained in the total quantity of light (V R + V C ) are different.
  • the ratio of V R or V C to (V R + V C ) is determined accord­ing to this method, a symmetry is obtained with respect to the achromatic axis Z, as shown in Fig. 3, so that no divergence results, as different from the above.
  • the ratio V C /(V R + V C ) is used as the color difference axis, the individual colors can be clearly extracted as follows: for the red colors: 0 ⁇ V C /(V R + V C ) ⁇ 0.5; for the black colors: V C /(V R + V C ) ⁇ 0.5; and for the cyan colors: 0.5 ⁇ V C /(V R + V C ) ⁇ l.0 (9).
  • the color difference signal informa­tion can be expressed as follows: V R /(V R + V C ); or V C /(V R + V C ) (l0).
  • the ranges of the black, cyan and red colors are determined on the basis of the aforementioned color extraction method.
  • An ideal achromatic copying tool would be positioned on the achromatic axis Z as shown in Fig. 3.
  • the copying tool is within a region having a width in the neighborhood of 0.5 with respect to the ratio V C /(V R + V C ) because of various black copying tools and originals are present.
  • the black colors belong the region having a small value with respect to the value (V R + V C ).
  • the black (or anchromatic) color range basically takes a T-shaped form, as shown in Fig. 4, although it has more or less irregularities.
  • the hatched region belongs to the black (or achromatic) range.
  • the image scanned and read optically is coded in a multiple-valued notation with a predetermined threshold value so that it may be outputted and expressed.
  • a predetermined threshold value such as the luminance level (or reflectivity) of the original but can be handled more easily by using a value which corresponds to the reflection density of the original. It is, therefore, necessary to clear the cor­respondences between the values (V R + V C ) and the re­flection densities of the original for the individual black, cyan and red color ranges.
  • the correspondence between the luminance signal informa­tion (V R + V C ) and the reflection density is made for each range of the black, cyan and red colors.
  • the density measured results are com­pared with the various charts which have their colors separated and plotted on maps.
  • the density cor­respondences are expressed in 4 bits (or at l6 levels) of 0 to 2.0. If the l6 levels are coded in a hexa­decimal notation, the density correspondences are shown for the individual color ranges in Fig. 5.
  • the color extraction maps (i.e., the density cor­respondences in a ROM) corresponding to the color ranges shown in Fig. 4 are prepared on the basis of the density correspondences shown in Fig. 5 and are presented in Fig. 6. Thick lines appearing in Fig. 6 represent the boundaries of the false color range regions.
  • Figs. 7 and 8 are block diagrams showing examples of the structure of the image processing circuit of the prior art based on the color extraction maps shown in Fig. 6.
  • An optical image is separated into a red channel and a cyan channel by a color-separ­ating optical system (although not shown).
  • These red and cyan channels are then photoelectrically transduced by a CCD l and a CCD 2 and are amplified by the ampli­fiers 3l and 32 of an amplification unit 3, respec­tively.
  • the signals thus amplified enter an A/D conver­sion unit 4 and are converted into digital data, respectively, by A/D converters 4l and 42.
  • the output V R of the A/D converter 4l and the out­put V C of the A/D converter 42 enter color-extracted information preparing means 5 and are given as addresses to a (V R + V C ) memory 5l and a [V C /(V R + V C )] memory 52, respec­tively.
  • the (V R + V C ) memory 5l outputs the (V R + V C ) data corresponding to the input address
  • the [V C /(V R + V C )] memory 52 outputs the [V C /(V R + V C )] data corresponding to the input address.
  • These data enter color information storing means 6 and are given as addresses to a red/cyan memory 6l and a black memory 62, respectively.
  • red/cyan memory 6l and the black memory 62 are stored with the density correspondences for the in­dividual color ranges of the color extraction map shown in Fig. 6 so that the density values corresponding to the input addresses are outputted and stored in buffers 7 and 8.
  • the density correspondences stored in these buffers 7 and 8 are selected by a color select circuit l0 which is made receptive of color select signals B, B and R.
  • the circuit structure shown in Fig. 8 is similar to that of Fig. 7 except that the memories for storing the density correspondences are divided for the indivi­dual color ranges such as a red memory 63, a black memory 62 and a cyan memory 64 so that buffer memories 7, 8 and 9 are accordingly added.
  • the addresses of the color extracting ROM are different by the third term of the equation (l4) and the second term of the equation (l5).
  • both the addresses of V R + V C and V C /(V R + V C ) will change in the increasing direction if the original image is in a red color, and the addresses V C /(V R + V C ) will change in the decreasing direction if the original image is in a cyan (or blue) color.
  • Fig. 9 is a diagram showing the positional changes in the ROM table at this time.
  • the output which should be in the series of white - black - black - white, will be in the series of white - blue - black - red - white, as shown in Fig. l2(A) (wherein W represents "white"), when two black lines are picked up.
  • W represents "white"
  • the output fluctua­tions are little influenced so that the color ghosts are hard to result.
  • the color ghosts are generated such that the output is from the series of white - black - black - red - white to the series of white - blue - black - black - ­red - white.
  • the line patterns of red and blue colors are similar, as shown in Fig. l2(B) and l2(C).
  • Fig. l3 illustrates a color ghost image in which a pixel displacement of l/2 pixel is caused.
  • region a represents an intrinsic black region
  • region b represents a red ghost region
  • region c represents a blue ghost region.
  • the focal points of the individual colors R, G and B be located at the same position.
  • the focal points are generally different for the individual colors.
  • the three colors R, G and B i.e., C, E and F lines
  • two could have an identical focal point, but the remaining one would be different.
  • the focal point could be aligned with the optical axis but would gradually deviate at a large height of the image.
  • the magnification factor becomes so different that the pixel displacement will appear effectively the larger for the position the closer to the edge of the CCD.
  • the causes described above will generate the color ghosts, as shown in Fig. l4.
  • the reflected beam (or optical image) of an original document 95 is condensed by a lens 96 and focused on a CCD 97.
  • let­ter A represents a normal focal point
  • letter B a CCD position mainly for the focusing purpose.
  • Broken curve C indicates the focal plane of the lens
  • let­ters L1 and L2 indicate the normal length and the length mainly for the focusing purpose, respectively.
  • Figs. l5 are diagrams showing the states in which the color ghosts are generated.
  • Fig. l5(A) illustrates the state of the color ghosts in case the sudden noises are superposed
  • Fig. l5(B) illustrates the state of the color ghosts in case the low-frequency noises are superposed.
  • the sudden noises are generated in the solid black portion to gen­erate the dotted blue color ghosts.
  • the color ghosts are generated in the solid black portion over a (hatched) wide red range.
  • the color ghosts are generated for the various causes at the color image processing step.
  • the countermeasures for the color ghosts have been tried in various manners in the prior art. Most of the countermeasures for the color ghosts in the prior art are devoted to binary image data after the color extraction but cannot be applied to an image data having a pixel of 6 to 8 bits, for example.
  • the countermeasure of the prior art has a drawback that it is difficult to judge what color makes the ghost image if the color number increases to as many as three to seven.
  • the invention disclosed in Japanese Patent Laid-Open No. l28872/l984 removes the color ghosts by storing in a corresponding manner the combina­tion of the colors of pixels around a pixel noted and the inferred color of the noted pixel and by logically operating the informations of the color of the noted pixel and the inferred color to determine the color of the noted pixel.
  • the image data are in two binary colors, and the logical operation prefers the black color because of necessity for clearing the black image.
  • Figs. l6 show examples in which the color ghosts are generated in this case. This example is di­vided into three types, as shown in Figs. l6(A) to l6(C), in all of which the center represents the intrinsic line width.
  • the color ghosts may appear only at one side of the line or may not appear in the least (that is, they are generally reluctant to appear at the red edge).
  • the color ghosts always have a possibility of appearance (depending naturally on how to determine the binary threshold value).
  • the image processing methods of the prior art cannot output individual color signals (or density data) simultaneously so that the color ghosts cannot be removed by the single original scanning.
  • the present invention has been conceived in view of the description thus far made and has an object to provide an image processing method, which is enabled to code a multiple-valued notation on real time for each color with an identical circuit, and an image processing apparatus having a simple construction.
  • the image processing apparatus for solving the aforementioned problems in accordance with the present invention is characterized in that the image of an orig­inal is processed by picking up the image, by extracting the colors of the image, and by subsequently composing the image information of a color designating information and a density information.
  • Another object of the present invention is to pro­vide an image processing apparatus which is enabled to ensure removal of color ghosts with a simple consturc­tion.
  • This object can be achieved by simultaneously con­ducting a color ghost compensation after color extrac­tion and an image processing in an image processing apparatus which obtains a plurality of color signals by picking up an image and extracting it into colors.
  • an MTF correction i.e., a correction for enhancing a resolution
  • MR magnification
  • a gradation a gradation
  • the MTF correction can be conducted after the color ghost correction. If the image processing time is considered, however, the density information is retained at the ghost correction, and a time loss is involved despite the processing is possible. Therefore, the par­allel processing, in which the color ghost processing and the MTF corrections are conducted simultaneously in parallel, is conducted so as to satisfy the speedup of the processing time and the high qualification at the same time.
  • the aforementioned object can be achieved by con­ducting a color ghost correction after a color extrac­tion and processing a corrected image infor­mation in an image processing apparatus for picking up an image and separating it into colors to obtain a plu­rality of color signals.
  • the color ghost correc­tion is conducted while storing the image data as multiple-valued data.
  • the aforementioned object can be achieved by giving pixel information as color designating color codes and a density information when in a color extraction and by making a ghost correction with the color codes after the color extraction in an image processing apparatus for obtaining a plurality of color signals by picking up an image.
  • one pixel information is given when in the color extraction with the color des­ignating information and the density information so that the ghost correction is conducted after the color extrac­tion with the color designating information.
  • the aforementioned object can be achieved by giving the pixel information in color designating color codes and a density data when in a color extraction and by making a color ghost correction to the color codes after the color extraction in case an image is picked up to obtain a plurality of color signals.
  • the ghost correction is conducted to the color codes by giving one pixel infor­mation as the color designating color codes and the den­sity data when in the color extraction.
  • Still another object of the present invention is to provide an image processing apparatus which is enabled to obtain a high quality image by ensuring a color ghost removal by a single original scanning.
  • This invention can be achieved by an image process­ing apparatus for obtaining a plurality of color signals by picking up an image, which apparatus includes means for giving one pixel information as color designating color codes and a density data and separating the pixel information into the color designating color codes and the density data, and means for conducting a color ghost processing on the basis of the color designating color codes.
  • the pixel informations are stored for the individual color ranges in different memories, which are commonly accessed to when in the color ghost removal on the basis of an image pickup signal.
  • a further object of the present invention is to provide an image processing apparatus which is enabled to store multi-color image informations in memories by a single image pickup operation.
  • This object can be achieved by composing a pixel information from a color designating information and a density information when in a color extraction, by des­ignating the storage address of a memory with the color designating information and a counted value of a posi­tion of each pixel, and by storing the corresponding density information, in an image processing apparatus for picking up an image and extracting it into colors and for subsequently storing the extracted colors as image data.
  • a color extraction map is prepared giving one pixel information in color designating color codes and density data.
  • the color extraction of the prior art is conducted by the circuit shown in Fig. 8.
  • the red memory and the cyan memory that of the black memory, for example, stores the data (e.g., the data of one pixel and four bits in the case of Fig. 8) only in the T-shaped region, as shown in Fig. 6, whereas the data in the other regions are 0. Similar storages are conducted in the red memory and the cyan memory.
  • color des strictlyignating color codes i.e., color designating informa­tions
  • density corresponding values i.e., density informations
  • these density data and color des­ignating informations are combined into the image data at the unit of one pixel.
  • the density corresponding value having a value D in Fig. 6 changes as follows:
  • a white code corresponding value "30" is written in the regions other than the T-shaped region.
  • a color extraction memory having each color region stored with image information is prepared in advance.
  • color extraction memories in a number equal to that of the color extractions to store the image informations (or color-separated data) of the individual color ranges.
  • black, red and blue color extraction memories are prepared black, red and blue color extraction memories.
  • Figs. l8 illustrate the storing states of individ­ual color ranges in the memories.
  • Fig. l8(A) shows the storing state of the black memory
  • Fig. l8(B) shows the storing state of the red memory
  • Fig. l8(C) shows the storing state of the cyan memory.
  • letter X designates the density data (or the den­sity corresponding value).
  • the images of an original, which have their colors separated into red and cyan colors, are picked up by CCDs.
  • the original image is illuminated by a light source, and its reflected light is separated into red and cyan optical images by a color separating optical system such as a dichroic mirror. And, these optical images are picked up by CCDs (i.e., red CCD and cyan CCD) dis­posed for the individual red and cyan colors.
  • CCDs i.e., red CCD and cyan CCD
  • the resultant signals are designated at V R and V C .
  • a luminance in­formation signal (V R + V C ) and a color difference infor­mation signal V C /(V R + V C ) are obtained.
  • the step of generating the luminance information signal (V R + V C ) and the color difference information signal V C /(V R + V C ) on the basis of the image signals V R , V C will not be described in detail because they have been described hereinbefore.
  • the color separation memories are accessed to by addressing the data (V R + V C ) and V C /(V R + V C ) to obtain image data.
  • the color extrac­tion memories shown in Figs. l8 can be commonly ac­cessed to by addressing the luminance information signal (V R + V C ) and the color difference information signal V C /(V R + V C ). From the black memory, the red memory and the cyan memory, there are outputted the image data which are stored in the accessed addresses.
  • the image data outputted are a combination of the color des­ignating color codes and the density data.
  • the ghosts are corrected for the color codes of the image data obtained.
  • the appearances of the color of a pixel noted and the colors of peripheral pixels and the relations of the actual color of the noted pixel are examined to determine the color of the noted pixel uniquely.
  • N colors including the white color
  • M 5 to 7.
  • the processing described above is required to set the density data at 0 for the pixels which have changed from the data in the colors other than white to the white color.
  • the color ghost correction can be performed in any direc­tions including the main and auxiliary scanning direc­tions.
  • An image processing is conducted on the image data having their ghosts corrected.
  • the MTF correction for improving the resolution
  • the gradation conversion for changing the gradation
  • the image having been subjected to the image processing is coded in a multiple-valued notation and has its output displayed by a CRT and a printer.
  • Fig. 2l is a block diagram showing one example of an image processing apparatus for implementing the method of the present invention.
  • the same components as those of Fig. 8 are designated at the same reference numbers.
  • the circuit as shown, corrects the color ghosts in the main scanning direction but can be applied likewise to the auxiliary scanning direction if line memory are used.
  • the black memory 62, the red memory 63 and the cyan memory 64 are stored with the image data (i.e., the color-extracted data).
  • the image data are outputted. Then, for example, the color codes of more significant 2 bits and the density data of less significant 4 bits of the image data outputted from the individual memories are stored in latch circuits 90 and 9l, respectively. Then, the output of the latch circuit 9l enters a color pattern ROM 92 whereas the output of the latch circuit 90 enters a white information preparing circuit 93. Of these outputs, the density data are stored in the latch circuit 90 through an OR gate 80, and the color codes are stored in the latch circuit 9l through a gate 8l.
  • the ROM address is expressed as follows and is coded at 3EO in a hexadecimal notation:
  • the address coded at 3EO stores a code representing the black color, i.e., the data 0 so that the data (0, 0) are outputted from the color pattern ROM 92.
  • the density information latched by the latch circuit 90 is subjected to a predetermined processing by the subse­quent white information preparing circuit 93 until it is latched by the latch circuit 94. And, the density data latched by the latch circuit 90 and the color code out­putted from the color pattern ROM 92 are synchronously outputted.
  • the color code is corrected to correct the color ghosts.
  • the output of the white information preparing circuit 93 is controlled with the output of the color pattern ROM 92.
  • the color code (l, l) i.e., the density data of the white data is reduced to 0 whereas the data of other colors are outputted as they are. This prevents the dotted color ghosts from appearing in the white blank.
  • Those color patterns are preferably determined by stressing the black color, in which what is accepted as the color ghosts be changed into the black color, because many originals usually have black patterns on the white blank.
  • the present invention should not be limited thereto but can use a two-dimensional color pattern (which has a size of 3 ⁇ 3, for example).
  • the hatched region as shown in Fig. 22, presents a notable pixel.
  • the ghosts may have a width of about 2 pells to make it necessary to consider the colors of 2 pells at the two sides of the pixel noted.
  • the size required is as large as five pixels. This requirement makes it easy to use the one-dimensional color pattern.
  • the pat­tern number is drastically increased, as compared with the one-dimensional one, but the color information around the noted pixel is hard to introduce.
  • the color pattern may be either one-­dimensional as l ⁇ 7 (in which the scanning operations need be in the main and auxiliary directions, as shown in Fig. 23) or two-dimensional as 3 ⁇ 3 (in which the color pattern may be moved in one direction, e.g., in the main scanning direction).
  • Fig. l9(B) shows an example of the output color codes in the case of the color pattern of l ⁇ 7.
  • the present invention employs the fol­lowing method for coding a multiple-valued notation for each color on real time by the use of a common circuit.
  • Fig. 24 is a block diagram showing the structure of another embodiment of the present invention, especially its binary circuit.
  • the same components as those of Fig. 2l are designated at the same reference numerals.
  • Reference numeral 83 designates a binary circuit for coding the density informations from the OR gate 80 in a binary notation on the basis of the threshold data from a threshold ROM 82.
  • Numeral 84 designates a black memory plane for storing the data of the black color of the binary data coming from the binary circuit 83.
  • Numeral 85 designates a red memory plane for storing the data of the red color of the same, and numeral 86 des­ignates a cyan memory plane for storing the data of the cyan color of the same.
  • Numeral 87 designates a decoder for decoding the color codes or the B.B.R. codes (or the B.B.R. signal) coming from the gate 8l through a switch SWl to generate a select signal for selecting the individual memory planes 84 to 86.
  • Numeral 88 designates a color control circuit made receptive of the B.B.R. codes, the memory mode signal M and the color code signal.
  • Numeral 89 designates a gate controlled by said color control cir­cuit 88. The gate 89 receives either the output of the binary circuit 83 or any output of the memory planes 84 to 86 through a switch SW2 and sends its output as the output of the shown whole circuit to a recording device (e.g., a printer).
  • a recording device e.g., a printer
  • Numeral 95 designates a clock gen­erator for generating clock pulses and sending them as shift clocks to the CCDs l and 2 and as address clocks to the threshold ROM 82.
  • Numeral 96 designates a counter for counting the output clocks of the clock gen­erator 95 to feed the memory planes 84 to 86 with its output as addresses indicating the positional informa­tion. The operations of the circuit thus constructed will be described in the following.
  • the red and cyan optical images having their colors separated by a color separating optical system are photoelectrically converted for each of their pixels by the CCDs l and 2 and are then converted normalized digital data by the A/D converters 4l and 42.
  • This normalization can be realized by making such a full-scale adjustment that the output when the reflected light of a white density plate is incident may take a full-scale value.
  • the outputs V R and V C of the A/D con­verters 4l and 42 are fed as addresses to the (V R + V C ) memory 5l and the V C /(V R + V C ) memory 52, respectively.
  • the luminance signal data (V R + V C ) corresponding to the input address are outputted from the (V R + V C ) memory 5l, whereas the color difference signal data V C /(V R + V C ) corresponding to the input address are out­putted from the V C /(V R + V C ) memory 52. Both of these data are inputted commonly on addresses to the color extraction memories 62 to 64.
  • the red memory 63 and the cyan memory 64 there are outputted image data which correspond to the input address and contain the color designating informations and the density informations, as has been described hereinbefore.
  • the den­sity information are inputted to and outputted from the OR gate 80 until they enter the binary circuit 83.
  • This addition at the OR gate 80 raises no difficulty, because the individual color extraction memories 62 to 64 have such internal contents that all the data in the regions having no relation are at 0, as shown in Figs. l8.
  • the color designating informations enter the gate 8l and are outputted as the color codes from said gate 8l.
  • the color codes thus outputted from the gate 8l enter the ROM 82 to select a binary threshold value.
  • the color codes (00), (0l) and (l0) represent the black, blue and red colors, re­spectively.
  • the dither matrices of 2 ⁇ 2 shown in Fig. 25 are used for the individual colors.
  • the output (threshold) values of the dither matrices shown in Fig. 25 are determined, as shown in Fig. 25, for the input clocks CLK and the color codes.
  • the threshold values are outputted from the threshold ROM 82 for the individual color codes sequentially in synchro­nism with the clocks CLK.
  • the threshold data thus outputted enter the binary circuit 83 in which they are compared with the density informations inputted separately. Binary data based on the compared result are outputted from the binary cir­cuit 83.
  • the color codes selected by the switch SWl are decoded by the decoder 87 so that the memory planes of the individual colors are selected in accordance with said decoder output.
  • the position of each pixel can be located by counting the output clocks of the clock generator 95.
  • the counter 96 gives the positional information of each pixel as the address to each of the memory planes 84 to 86.
  • the output (i.e., the binary data) of the binary circuit 83 is stored at the address thus designated.
  • the operations described above can be made by the single image pickup action.
  • only one color can be selectively outputted by arithmetically operating the B.B.R. codes and the color codes.
  • the switches SWl and SW2 By switching the switches SWl and SW2, moreover, the individual color informations stored by the single action can be outputted one by one by controlling the B.B.R. codes.
  • the memory mode selection signal M is used.
  • the color control circuit 88 is constructed to output the signals shown in Figs. 27 in response to the B.B.R. codes, the color codes and the memory mode selection signal M.
  • the gate 89 is opened for the output at "l".
  • the gate 89 is opened only when the B.B.R. codes and the color codes are coincident.
  • the gate is open at all times. In the memory mode, therefore, the information of the memory plane designated by the B.B.R. codes is outputted.
  • the image informations are composed of the color designating informations and the density informa­tions, and the threshold value for coding the density informations in the multiple-valued notation is selected by the use of those color designating informations.
  • the colors are limited to the three, i.e., the black, red and blue (cyan) colors, to which the present invention should not be limited.
  • the multiple-valued notation be limited the binary notation disclosed in the embodi­ment, but present invention can make a multiple-valued notation such as ternary or quarternary notation.
  • the present embodiment employs the dither matrix of 2 ⁇ 2, but the present invention should not be limited thereto.
  • the threshold ROM is selected with the color codes, but it is apparent that the selection of the threshold ROM can be accomplished by the color con­trol circuit.
  • the three black, red and cyan memories are used to store the data shown in Figs. l8, but similar results can be obtained by the use of a single memory.
  • the phenom­enon called the color ghosts appears in case the color original is to be picked up by employing a plurality of pickup elements.
  • the color ghosts can be removed by processing the color designating informa­tions. This makes it possible to conduct the color ghost removal after the color extraction and then to store the informations of plural colors by the single image pickup action using the circuit example shown in Fig. 24.
  • a variety of image processings can also be con­ducted on the density informations before the multiple­valued notation.
  • an image processing system of a simple con­struction which is enabled to conduct the multiple-­valued notation for each color on real time by a common circuit, can be realized by using the density informa­tions and the color designating informations as the color extracting image data.
  • the color extraction ROM is exemplified by the plural memories. It is, how­ever, apparent that the circuit can be realized by a single ROM if the color codes are used.
  • Fig. 28 is a block diagram showing a further em­bodiment of the present invention.
  • the same components as those of Fig. 2l are designated at the same reference numerals.
  • the color designating in­formations are sent through the gate 8l to the latch circuit 9l.
  • This latch circuit 9l is constructed of latches of 5 bits, the outputs of which are inputted as addresses to the color pattern ROM 92.
  • the density informations are sent through the OR gate 80 to the latch circuit 90.
  • the outputs of this latch circuit 90 are fed to an MTF correction ROM 97, the output of which is fed to the white information preparing circuit 93.
  • This white information preparing circuit 93 is fed with not only the output of the MTF correction ROM 97 but also the output of the color pat­tern ROM 92. And, the output of said white information preparing circuit 93 is fed as the density informations to the latch 94. The operations of the apparatus thus constructed will be described in the following.
  • the latch circuit 90 is caused to latch the density informations by the action similar to that of Fig. 2l.
  • the density informations thus latched are inputted as addresses to the subsequent MTF correction ROM 97.
  • This MTF correction ROM 97 outputs the MTF corrected data.
  • the MTF correction is one for improving the resolution.
  • the output amplitude of a CCD drops to change the apparent density.
  • This disadvantage is corrected by the MTF correction.
  • the state of a pixel noted is discriminated from the states of the pixels around the noted one to make a correction in a direction to enhance the resolution.
  • the white information preparing circuit 93 receives the MTF corrected data.
  • the color pattern ROM 92 receives the color codes of 5 bits of the noted pixel and its peripheral pixels to output the corrected color codes for removing the color ghosts. These color codes are inputted to the latch 94 and to the white information preparing circuit 93. And, this white information preparing circuit 93 sets all the out­puts of the MTF correction ROM 97 at 0, when the color codes are (l, l) representing the white color. In other color codes, the output of the MTF correction ROM 97 is sent as it is as the density informations to the latch 94. The density informations and color codes thus latched by the latch 94 are outputted in synchronism with each other.
  • the color ghost processing and the MTF correction processing are conducted in par­allel.
  • the not-shown timing control circuit controls the read timings of the color pattern ROM 92 and the MTF correction ROM 97.
  • the one-dimensional processing is not popular, as in the present embodiment, but two-­dimensional processing is preferable for a higher image quality. In this latter case, it is quite natural that the MTF correction can be made two-dimensionally by the use of a line memory.
  • the correction is made by the use of the ROM, but the present invention may be modified such that the image is picked up by calculating the quantity of correction with a CPU in advance by the use of the ROM and by writing the calculated data in a RAM. It is also apparent that a digital filter or the like can be used.
  • the description thus far made is directed, by way of example, to the case in which the MTF correction is conducted as the image processing to be conducted in parallel with the color ghost processing.
  • the present invention should not be limited to that case, but the MTF correction may be replaced by the image em­phasis or gradation changing processing.
  • the color ghosts can be prevented without fail from appearing by conducting the color ghost processing in parallel with the processing of the image informations after the color extraction.
  • the table addresses for the color separation are common among the individual colors so that a simple apparatus can be realized.
  • Fig. 29 is a block diagram showing the structure of a further embodiment of the present invention.
  • the same components as those of Fig. 2l are designated at the same reference numerals.
  • reference nu­meral ll0 designates a color ghost removing circuit for removing the color ghosts in response to the outputs of the color extraction memories 62 to 64.
  • Numeral l08 designates an arithmetic circuit for performing an arithmetic operation in response to the color codes of said color ghost removing circuit ll0 and the color select signal B.B.R.
  • Numeral l09 designates a gate cir­cuit for gating and outputting the density information output of the color ghost removing circuit ll0 in re­sponse to the output of the arithmetic circuit l08.
  • the black memory 62, the red memory 63 and the cyan memory 64 output image data corresponding to the input addresses and feed them to the color ghost removing cir­cuit ll0.
  • This color ghost removing circuit ll0 con­ducts the color ghost removal by the color pattern method, which has been described with reference to Fig. 2l, to output the density informations and the color codes corresponding to the former.
  • the arithmetic cir­cuit l08 outputs the "l" level, only when the color codes and the color select signal B.B.R. are coincident, and otherwise the "0" level.
  • the gate circuit l09 is opened to pass the density informations therethrough, only when the output of the arithmetic circuit l08 is at "l", and is closed when the same out­put is at "0".
  • Fig. 30 is a diagram showing one example of the construction of the arithmetic circuit l08.
  • Both the color codes and the color select signals B.B.R. (which will be called the "B.B.R. codes") are individually com­posed of 2 bits.
  • the color codes and the B.B.R. codes have their more significant bits fed to a first exclu­sive OR gate G1 and their less significant bits fed to a second exclusive OR gate G2. And, both the outputs of those first and second exclusive OR gates G1 and G2 are fed to an AND gate G3, the output of which is inputted to the gate circuit l09. Only when both the color codes and the B.B.R.
  • Fig. 3l is a diagram tabu­lating the outputs of the arithmetic circuit shown in Fig. 30.
  • Fig. 32 is a circuit diagram showing one example of the multiple-valued coding circuit.
  • the threshold ROM 82 is fed with the color codes, a level setting signal and the clocks CLK to output optimum threshold data ac­cording to the color codes and level setting signal in­putted.
  • the density informations and the threshold data are latched by a latch l04 and then are compared by a subsequent comparator ll2 so that they are converted into binary data. Since the levels are set for the in­dividual colors in the shown circuit, it is possible to set the density levels independently of the colors.
  • the description made above is directed to the binary case, the circuit can be applied for the ternary or quarternary coding. In this application, the thresh­old ROMs, the comparators and the latches are increased so that two or more threshold values may be compared.
  • Fig. 33 shows the positional re­lationships between the locations (as indicated at circle ⁇ ) of the original lattice point and the loca­tions (as indicated at letter X) of the size-changed lattice point for the magnification of 4/3.
  • Fig. 34 letters J and K represent the positions of an original whereas P J and P K represent the density data (or informations) at those points, and the grada­tion is assumed to have 8 levels.
  • the density data at the position (as indicated at X) of the size-­changed lattice point are determined to 3 by the cor­responding point on a straight line joining P J and P K . In this way, a size change can be accomplished.
  • the locations of the size-changed lat­tice point, the density data and so on can be retrieved at a high speed by adopting the table look-up type of the ROM.
  • Fig. 35 is a block diagram showing one example showing a size changing circuit.
  • the density informa­tions are latched by latches l2l and l22 in response to a clock CLK0 and fed as addresses to an interpolation ROM l23.
  • the size changing signal is fed to a ROM selector, which outputs two ROM select sig­nals #l and #2 in depending upon whether the magnifica­tion is larger (for enlargement) or smaller (for reduc­tion) than l.
  • An enlargement position ROM l25 is selected in response to the signal #l whereas a reduction position ROM l26 is selected in response to the signal #2.
  • clocks CLKl are fed to a counter l27, which counts the number of the clocks CLKl to feed its counted value as an address to the enlargement posi­tion ROM l25 and the reduction position ROM l26.
  • these ROMs l25 and l26 output both the data of the counter l27 and the data latched in the address corre­sponding to the magnification signal to feed them to a latch l28.
  • these data are latched by said latch l28.
  • the information concerning the position data is fed as an address from the latch l28 to the interpolation ROM l23, which out­puts as interpolation data the data which are latched in the addresses corresponding to the outputs of the latches l2l, l22 and l28.
  • interpolation data are inputted to and latched by a latch l29 in response to the clocks CLKl until they are outputted to a binary circuit (although not shown).
  • the titled processings are not necessary for ordi­nary purposes but are used for intentionally making lines thicker or thinner or removing smudge from the surface of the original.
  • the present processing may be accomplished by providing a special circuit but can be achieved by devising the color pattern using the color ghost processing circuit.
  • the line width of a writing tool is about 0.l to 0.2 mm and takes 2 to 3 dots when the image is picked up at l6 dots/mm.
  • a color pattern of l ⁇ 7, the center black pell corresponding to the following pattern may be treated as a white pell: White-White-White-Black-White-White-White.
  • the noise color should not be limited to the black.
  • the line can be thinned by reducing the two black pixels to one, as follows: White-White-White-Black-White-White-White.
  • the line can be thicked by increasing the two black pixles to three, as follows: White-White-Black-Black-Black-White-White.
  • the pre­sent invention can be applied to a two-dimensional pro­cessing.
  • the line can be thinned by changing the black of the (center) pixel noted into white for the patterns shown in Fig. 36.
  • the line can be thicked by changing the black of the (center) pixel noted into black.
  • these processing should not be limited to the dichroic patterns but can be extended to other colors.
  • the emphasis of an image is a technique to be used for emphasizing a line portion or a portion where den­sity abruptly changes. If a filter of l ⁇ 3 of [-l, 3, -l] is applied in the main scanning direction, for ex­ample, only the edge in the main scanning direction is emphasized. If each density data is inputted as an address to the ROM which has latched an input informa­tion and made a convolution arithmetic of it in advance, the output of the ROM is an image-emphasized signal.
  • an input signal is latched by latches l3l and l32 to prepare three differ­ent input signals I1, I2 and I3, which are fed as addresses to a ROM l33.
  • This ROM has made the following calculation in advance: I1 ⁇ (-l) + I2 ⁇ (3) + I3 ⁇ (-l). If, at this time, the density data of a point noted and the two sides are inputted as the addresses of that ROM, an image emphasis signal is outputted from the ROM l33.
  • this filter of l ⁇ 3 is applied to the auxiliary scanning direction, the edge in this direction is apparently emphasized. Moreover, this emphasis can also be extended to a two-dimension. For example, an arithmetic of 3 ⁇ 3 may be conducted by using the following data: On the other hand, generally speaking, the filter should not be limited to the above-specified one but can be varied in accordance with the using purpose.
  • This modification employs a circuit shown in Fig. 39.
  • the input data are sequentially delayed by latches L11 to L32 and line memories LM1 and LM2 so that they are fed as addresses 3 ⁇ 3 to a ROM l40.
  • a combination of the informations of a noted pixel and its periphery is calculated and stored in advance in the ROM l40.
  • the content of the ROM l40 is addressed to generate an output in accordance with the informations of the noted pixel and its periphery.
  • Fig. 39 shows an example using the filter of 3 ⁇ 3.
  • the filters of n ⁇ n it is quite natural that if a filter of n ⁇ n is used the numbers of the line memories and the latch circuits may be increased or decreased.
  • the aforementioned one-dimensional filter can be applied first to the main scanning direction and then to the auxiliary scanning direction.
  • the example described above has employed the ROM, which may be an EPROM or EEPROM when in a low-speed operation and a bipolar ROM when in a high-speed opera­tion.
  • the same purpose can be achieved by using a RAM in place of the ROM, by feeding the RAM with data calculated in a CPU or data from a ROM and by addressing this RAM to produce an output. More strictlyover, a digital filter may be used for the purpose.
  • a variety of color originals are to be picked up into images, many of which have a narrow density range.
  • the gradation change is intended when a clearer image is to be obtained from such images.
  • Fig. 40 illustrates an example of gradation changing characteristics.
  • the abscissa indicates the density of an original image
  • the ordinate indicates the change density. If the density range of the origi­nal image is designated at W1 and if the gradation change is conducted with a curve f of Fig. 40, the changed image has a density range of W2. Then, W2 > W1 so that an effect similar to the histogram equalization is obtained.
  • the look-up method of the ROM table may be adopted.
  • the curve f (generally, a variety of curves are prepared and used by switching their shapes) is stored in a ROM l4l, as shown in Fig. 4l, and the density of an original image is inputted as an address to the ROM l4l so that a changed density value may be outputted.
  • a desired curve may be selected from the change curves, as shown, by inputting a change curve designating signal.
  • the gate circuit l09 may be opened or closed depending upon whether or not the operation is achromatic.
  • a block defined by broken lines in Fig. 42 is an achro­matic circuit, the output of which is fed to the gate circuit l09.
  • the achromatic circuit is composed, as shown, of exclusive OR gates G11 and G12, a NAND gate G13 and a 3-state buffer G14, the output of which is fed to the gate circuit l09.
  • the gate circuit l09 When the achromatic signal is at "0" for the 3-­state buffer G14, the gate circuit l09 receives a "l" signal to output a density information. When the achro­matic signal is at “l”, on the contrary, the gate cir­cuit l09 receives the "0" signal so that it is closed. Then, no density information is outputted to effect the achromatism.
  • the ROM table In case the ROM table is used, on the other hand, total six data of (three) erasures of one color and (three) erasures of two colors may be stored in the ROM to output the "l"/"0" signals in accordance with the color codes.
  • a ROM to which is inputted as an address a code prepared for a 4-bit input, as shown in Fig. 43.
  • the ROM may change 2-bit color codes inputted into the output codes, as shown in Fig. 44.
  • Fig. 45 is a block diagram showing an example of the color changing circuit.
  • Reference numeral l50 de­signates a color change ROM for effecting the aforemen­tioned color change, which outputs the new color codes shown in Fig. 44 when it receives the color codes (of 2 bits) and the color change codes (of 4 bits). These new color codes are fed to the arithmetic circuit l08. The subsequent operations are similar to those of Fig. 29.
  • the description thus far made employs the color pattern method for the color ghost processing, to which the present invention should not be limited but can em­ploy another kind of color ghost processing method.
  • the gate circuit was used as the arithmetic circuit l08 but may be re­placed by the ROM table.
  • the color ghosts can be reliably removed with the simple construction by making the color ghost cor­rection for the image informations after the color extrac­tion and by subsequently conducting the image proces­sing. Because of the binary image informations, the method of the prior art has failed to process the image data after the color ghost processing.
  • a multiple-valued image is subjected to the color ghost processing so that its data can be subsequently processed.
  • the image is processed after the color ghost processing to invoke no deterioration of the image quality when in the MR (magnification) operation.
  • Fig. 46 is a block diagram showing the structure of a further embodiment of the present invention and an image processing system including a color ghost proces­sing circuit.
  • the same components as those of Fig. 28 are designated at the same reference numerals.
  • a block enclosed by broken lines in Fig. 46 is the color ghost removing circuit which is composed of the gates 80 and 8l, the latches 90 and 9l, the color pattern ROM 92 and the white information preparing circuit 93.
  • the color ghost removing circuit which is composed of the gates 80 and 8l, the latches 90 and 9l, the color pattern ROM 92 and the white information preparing circuit 93.
  • the white information preparing circuit 93 Of the out­puts of the color extraction memories 62 to 64, all the density informations are fed to the gate 80, the output of which is fed to the first latch circuit 90, the out­put of which in turn is fed to the white information preparing circuit 93.
  • the second latch circuit 9l employs a plurality of stages of shift registers, for example, so as to attain the color informations of a noted pixel and its peripheral pixels.
  • the output of the color pattern ROM 92 is fed to the white information preparing circuit 93 and the latch 94, whereas the output of the white information preparing circuit 93 is fed to the latch 94.
  • an image pro­cessing system which is enabled to remove the ghosts without failure by giving one pixel information in the color designating color codes and the density informa­tions and by converting the color designating data after the color extraction by the use of those color codes.
  • Fig. 47 is a block diagram showing the structure of a further embodiment of the present invention. Numerals attached to the signal lines, as shown, represent the bit numbers of the signals.
  • the optical images having their colors separated into red and cyan colors by a color separating optical system (although not shown) are photoelectrically converted by CCDs 28l and 282, respec­tively, and the resultant electric signals are amplified to predetermined levels by subsequent amplifiers 283 and 284.
  • the image signals thus amplified are converted into digital data by A/D converters 285 and 286, respectively.
  • These A/D converters 285 and 286 are so subjected to a full-scale adjustment in advance that they may take rated outputs when they receive the image pickup signal of a white reference density plate.
  • the data V R and V C outputted from the A/D con­verters 285 and 286 are normalized.
  • V R and V C are commonly inputted as addresses to a (V R + V C ) ROM 287 for storing luminance signal data and a V C /(V R + V C ) ROM 288 for storing color difference signal data.
  • the (V R + V C ) data and V C /(V R + V C ) data corresponding to the input addresses are out­putted from the (V R + V C ) ROM 287 and the V C /(V R + V C ) ROM 288, respectively, and are latched by latches 289 and 290, respectively.
  • the data thus latched by the latches 289 and 290 are commonly inputted as addresses to a black extraction ROM 29l, a red extraction ROM 292 and a blue extraction ROM 293.
  • the CCD 28l and CCD 282 correspond to the CCD l and CCD 2 of Fig.
  • the amplifiers 283 and 284 to the amplifiers 23l and 232; the A/D converts 285 and 286 to the A/D converters 24l and 242; the (V R + V C ) ROM 287 to the (V R + V C ) memory 25l; the V C /(V R + V C ) ROM 288 to the V C /(V R + V C ) memory 252; the black extraction ROM 29l to the black memory 262; the red extraction ROM 292 to the red memory 263; and the blue extraction ROM 293 to the cyan memory 264.
  • the black extraction ROM 29l, the red extraction ROM 292 and the blue extraction ROM 293 extract the color extraction map described with reference to Fig. 6 into a black (or achromatic) region, a red region and a blue region, respectively, to store the density data (corresponding to the densities) independently for the individual color ranges.
  • the aforementioned color designating color codes are stored in addition to the density data for each pixel in a manner to couple with the density data.
  • These ROMs 29l to 293 are stored with dummy data (at 0) in their addresses other than those assigned to them­selves.
  • the image data thus outputted from the individual color extraction ROMs 29l to 293 are separated into the color designating color codes and the density data, which are then processed separately.
  • the color codes are fed through a first composer 294 to a color ghost removing circuit 295, whereas the density data are fed through a second composer 296 to the color ghost remov­ing circuit 295.
  • the color ghost removing circuit 295 con­ducts such ghost removing operations (as will be de­scribed in detail in the following) in both the main and auxiliary scanning directions as will convert the color codes or ghosts into color codes (i.e., color codes matching the color of the circumstance) in another color, and outputs the color codes and density data, from which the ghosts are removed.
  • the color codes and density data thus outputted from the color ghost removing circuit 295 are latched by a latch 297.
  • the color codes are fed to a first decoder 298 so that they are decoded into four color designating signals of black, red, blue and full-output.
  • a color select signal i.e., B.B.R. signal
  • B.B.R. signal given separately is fed to a second decoder 299 and is demodulated into four color designating signals of black, red, blue and full-output.
  • the signal couples in the same colors are fed to individual OR gates 300 so that they are outputted therefrom as a black select sig­nal, a red select signal, a blue select signal and a full-output select signal, respectively.
  • the select signals thus outputted from the OR gates 300 in these signals are fed to a density signal separ­ator 30l.
  • the density data latched by the latch 297 have already been fed to said density signal separator 30l.
  • the density signal separator 30l selective output the den­sity data in the individual colors (as will be described in detail in the following).
  • the density data thus outputted from the density signal separator 30l are com­pared in a comparator 302 with a threshold value out­putted from a threshold circuit 303 so that they are converted into binary data. These binary data are out­putted as the image data.
  • the multiple­valued notation should not be limited to the binary one but can be extended to the ternary or quarternary one.
  • Fig. 48 is a circuit diagram showing an example of the detailed structure of the color ghost removing cir­cuit 295.
  • the circuit as shown, is enabled to conduct the color ghost removal in both the main and auxiliary scanning directions.
  • the color codes (of 2 bits) are fed to a first shift register 20l so that they are shifted in response to a data clock.
  • the shift register 20l outputs parallel data of 2 ⁇ 7 bits which are de­layed by 7 bits at the maximum from the color codes of 2 bits.
  • These parallel data are fed to a ghost removing ROM 202 for removing the ghosts in the main scanning direction.
  • This ghost removing ROM 202 conducts the ghost removal by the aforementioned color pattern method.
  • the color codes of a noted center pixel are converted from the color of the noted pixel and the colors of three pixels before and after the noted pixel into the color codes freed from the color ghosts.
  • the color codes thus having their ghosts removed in the main scan­ning direction are latched by a latch 203.
  • the density data (of 4 bits) are fed to a second shift register 204 of 5 bits and converted thereby into density data of the noted pixel in response to a data clock until they are latched by the latch 203.
  • l-line signals outputted separ­ately are counted by a 7-line counter 205, from which is outputted a signal specifying a later-described l-line memory.
  • the output (of 3 bits) of the 7-line counter 205 is decoded by a decoder 206.
  • the signals which will be called the "memory select signals"
  • one group are inputted to gates G1 to G7 and l-line memories M1 to M7, another group being inputted to gates G11 to G17 and G21 to G27, and still another group being inputted to a data selector 207.
  • the color codes and density data thus latched by the latch 203 are controlled by memory selecting signals from the decoder 206.
  • the memory selecting signals from the decoder 206 select gates G1 to G7 sequentially every one line and a memory connected to the selected gate is made in writing state. Accordingly, each of (the memories M1 to M7 is made in writing state by one line among seven lines and generates written data by other lines. Similarly, gates G11 to G17 are connected to color code input and output lines of the memories, so that gates are operated to be closed only when the memories are in writing state. The gates G21 to G27 are connected to density data input and output lines of the memories, so that gates are operated to be opened so as to generate density data of the picture element noted.
  • the data selector 207 has such a role that the color coded data, of which auxiliary scanning order is varied every lines are rearranged by the output of the decoder 206.
  • the writing operations of the memories M1 to M7 are effected in the order of M1, M2, .... M6, M7, M1 .... .
  • the memory M2 is read by the address (l3 bits) in which the signal before 6 lines from the line for writing in the memory M1 is stored
  • the memory M3 is read by the address in which the signal before 5 lines from the line for writing in he memory M1 is stored and similarly the memory M4 is read by the address storing the signal before 4 lines
  • the memory M5 is read by the address storing the signal before 3 lines
  • the memory M6 is read by the address storing the signal before 2 lines
  • the memory M7 is read by the address storing the signal before l line.
  • the density data of the picture element noted at this state is data stored in the memory M4.
  • the color codes (of l4 bits) of the individual lines thus outputted from the data selector 207 are fed as individual addresses to a ghost removing ROM 208 in the auxiliary scanning direction so that the color ghosts of the noted pixel are corrected in the auxiliary scanning direction.
  • the color codes thus outputted from the ghost removing ROM 208 have their color ghosts removed in both the main and auxiliary scanning directions.
  • the color codes thus having been subjected to the color ghost corrections are once latched in a latch 209 together with the density data outputted from the gate G21, until they are outputted.
  • color data and density data of 7 lines are stored in the l-line memories M1 to M7 and are read out, if necessary, so that they are processed. Since a real image has a capacity of several hundreds lines, it cannot be processed by the line memories of the shown number. Therefore, the overlap operation is conducted to switch new density data from the last l-line memory.
  • Fig. 49 is a circuit diagram showing an example of the detailed structure of the density signal separator 30l.
  • Density data D0 to D3 outputted from the latch 209 (as shown in Fig. 48) are commonly inputted to data selectors DS1 to DS4 provided for the black, blue and red colors and the full-output.
  • the color codes and the color select signals are decoded by the decoders 298 and 299.
  • the correspondences between the color codes and color select signals, and the colors are as follows:
  • the paired signals outputted from the com­mon output terminals (e.g., the 0 outputs of the de­coders 298 and 299) are fed to OR gates G31 to G33.
  • the OR outputs of these OR gates G31 to G33 are fed as enable signals to the data selectors DS1 to DS3.
  • the data selector DS4 is connected directly with all the output terminals of the decoder 299.
  • the corresponding color select signal of the decoder 298 or 299 is dropped to "0" to enable the predetermined data selector DS1 to DS4 through the OR gates G31 to G33 (however, the data selector DS4 is enabled directly by the output of the decoder 299).
  • the density data are outputted from the enabled data selector and are fed to the subsequent comparator 302 (as shown in Fig. 47).
  • the color extraction memories can be commonly accessed to so as to obtain color-extracted data in individual colors, when an image is to be picked up, by the density data of individual color ranges based on the color extraction map in the color extraction memories together with the color codes.
  • Fig. 50 is a block diagram showing the structure of a further embodiment of the present invention.
  • the same components as those of Fig. 24 are designated at the same reference numerals.
  • the block memory 62, the red memory 63 and the cyan memory 64 are stored with indi­vidual data which has its pixel information composed of the aforementioned color designating and density infor­mations.
  • the density informations and the color desig­nating color codes are outputted from the color extrac­tion memories 62 to 64 which are commonly addressed by the (V R + V C ) memory 5l and the V C /(V R + V C ) 52.
  • the density informations are commonly inputted to the memory planes 84 to 86 of individual colors for storing the density data.
  • the color codes are decoded by the decoder 87 so that the decoded outputs are inputted as memory select signals to the memory planes 84 to 86 of the individual colors.
  • the color codes are composed of two bits (which can discriminate four colors), for example.
  • the black mem­ory plane 84 is addressed for the code "00"; the cyan memory plane 86 is addressed for the code "0l”; and the red memory plane 85 is addressed for the code "l0".
  • the counter 96 counts the shift clocks of the CCDs l and 2 and accordingly the addresses of the pixels in the main and auxiliary scanning directions and feeds them to the memory planes 84 to 86 so that the density informations are stored in the address of the memory plane designated by the output of the decoder 87.
  • the multi-color image informations can be stored in the memory planes by the single image pickup action without repeating these actions of the number of the colors.
  • the color ghosts can be removed by processing the color designating informations.
  • the informations of plural colors can likewise be stored by the single image pickup action by conducting the color ghost re­moval after the color extraction and by subsequently using the circuit shown in Fig. 50.
  • a variety of image processings can be conducted for the density informa­tions before the multiple-valued notation.
  • Fig. 5l is a block diagram showing the structure according to a further embodiment of the present inven­tion and an image processing apparatus including a color ghost processing circuit.
  • the same components as those of Fig. 50 are designated at the same reference numerals.
  • reference numeral ll0 designates the color ghost removing circuit which has been described with reference to Fig. 29.
  • the image informations of the in­dividual colors stored in the color extraction memories 62 to 64 are read out and subjected to the color ghost removal. After this, the density informations are stored in the individual memory planes 84 to 86 which are selected by the decoder 87.
  • the den­sity informations can be subjected after the color ex­traction (or after the color ghost removal) to a variety of image processings (such as size change, image empha­sis or gradation), and the color codes are then decoded so that both may be stored in the memory planes.
  • the image informations are composed of the color designating informations and the density informa­tions, and the storage address of the memories are des­ignated in accordance with the color designating infor­mations and the counted values of the positions of the individual pixels to store the corresponding density informations.
  • the multi-color image informations can be stored in the memories by the single image pickup action so that the image can be processed at a high speed.

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Facsimile Image Signal Circuits (AREA)
  • Color Image Communication Systems (AREA)
  • Image Processing (AREA)
EP87109332A 1986-06-30 1987-06-29 Bildverarbeitungsverfahren und Vorrichtung dazu Expired - Lifetime EP0251278B1 (de)

Applications Claiming Priority (14)

Application Number Priority Date Filing Date Title
JP61154672A JPH0628378B2 (ja) 1986-06-30 1986-06-30 画像処理方法
JP61154669A JPS6310885A (ja) 1986-06-30 1986-06-30 画像処理装置
JP61154667A JPS6310883A (ja) 1986-06-30 1986-06-30 画像処理装置
JP154669/86 1986-06-30
JP61154670A JPS6310886A (ja) 1986-06-30 1986-06-30 画像処理装置
JP154671/86 1986-06-30
JP154670/86 1986-06-30
JP61154668A JPS6310884A (ja) 1986-06-30 1986-06-30 画像処理装置
JP154668/86 1986-06-30
JP15467186 1986-06-30
JP154672/86 1986-06-30
JP154666/86 1986-06-30
JP61154666A JPS6310882A (ja) 1986-06-30 1986-06-30 画像処理装置
JP154667/86 1986-06-30

Publications (3)

Publication Number Publication Date
EP0251278A2 true EP0251278A2 (de) 1988-01-07
EP0251278A3 EP0251278A3 (en) 1990-02-28
EP0251278B1 EP0251278B1 (de) 1995-11-29

Family

ID=27566175

Family Applications (1)

Application Number Title Priority Date Filing Date
EP87109332A Expired - Lifetime EP0251278B1 (de) 1986-06-30 1987-06-29 Bildverarbeitungsverfahren und Vorrichtung dazu

Country Status (3)

Country Link
US (1) US4980758A (de)
EP (1) EP0251278B1 (de)
DE (1) DE3751614T2 (de)

Cited By (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP0270090A2 (de) * 1986-12-03 1988-06-08 Konica Corporation Farbbildverarbeitungsgerät
EP0350930A2 (de) * 1988-07-15 1990-01-17 Konica Corporation Farbbildverarbeitungsgerät
EP0350919A2 (de) * 1988-07-15 1990-01-17 Konica Corporation Gerät zur Farbbildherstellung
US4907078A (en) * 1985-09-11 1990-03-06 Ricoh Company, Ltd. Method for reproducing color images which are distinguishable in a document having only two colors
EP0357407A2 (de) * 1988-08-30 1990-03-07 Kabushiki Kaisha Toshiba Bildsignalverarbeitungsgerät zur Verwendung in einem Farbbildwiedergabesystem
EP0388877A2 (de) * 1989-03-22 1990-09-26 Konica Corporation Farbbildverarbeitungsgerät mit Bilddiskriminierungsmitteln
EP0395032A2 (de) * 1989-04-28 1990-10-31 Konica Corporation Farbbildverarbeitungsgerät
EP0396980A2 (de) * 1989-05-12 1990-11-14 Dainippon Screen Mfg. Co., Ltd. Verfahren zur elektronischen Bildverarbeitung von Linienumrandungen
EP0422602A2 (de) * 1989-10-12 1991-04-17 Hitachi, Ltd. Verfahren und Vorrichtung zur Farbbildaufzeichnung
US5016096A (en) * 1987-09-21 1991-05-14 Konica Corporation Apparatus for color processing specifically inside or outside a detected region
EP0458978A1 (de) * 1989-12-15 1991-12-04 Hitachi, Ltd. Vorrichtung und verfahren zum aufbewahren von bildern
EP0484890A2 (de) * 1990-11-06 1992-05-13 Dainippon Screen Mfg. Co., Ltd. Verfahren und Vorrichtung zur Erzeugung eines überlappenden Bildgebietes
US5206719A (en) * 1989-05-10 1993-04-27 Canon Kabushiki Kaisha Image processing apparatus including means for extracting an outline
US5268771A (en) * 1989-02-15 1993-12-07 Hitachi, Ltd. Image restoration processing system and method
EP0765075A3 (de) * 1988-06-20 1997-06-04 Canon Kk Bildwiedergabe- und Bildverarbeitungsgerät

Families Citing this family (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5132788A (en) * 1989-02-25 1992-07-21 Minolta Camera Kabushiki Kaisha Image processing apparatus for processing respective image data obtained by reading an outputting image signal corresponding to pixels forming the original image
US5208663A (en) * 1989-03-22 1993-05-04 Konica Corporation Color image processing apparatus provided with an image discrimination means
US5198898A (en) * 1990-10-10 1993-03-30 Fuji Xerox Co., Ltd. Data compressing system for compressing serial image data with color information
JPH04175065A (ja) * 1990-11-08 1992-06-23 Canon Inc 画像処理装置
JP3332398B2 (ja) * 1991-11-07 2002-10-07 キヤノン株式会社 画像処理装置及び画像処理方法
JP3517757B2 (ja) * 1994-04-27 2004-04-12 株式会社リコー 画像処理装置
KR100432240B1 (ko) * 2000-06-16 2004-05-22 다이니치 세이카 고교 가부시키가이샤 씨씨엠 계산시스템, 씨씨엠 계산방법 및 기록매체
US7570403B2 (en) * 2005-03-16 2009-08-04 Kabushiki Kaisha Toshiba Color image processing apparatus
KR102469426B1 (ko) 2015-10-27 2022-11-22 삼성전자주식회사 이미지 처리 장치 및 이의 동작 방법

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPS56166666A (en) * 1980-05-28 1981-12-21 Ricoh Co Ltd Color picture processor
JPS60160774A (ja) * 1984-01-31 1985-08-22 Fuji Xerox Co Ltd ２色原稿読み取り装置
EP0270090A2 (de) * 1986-12-03 1988-06-08 Konica Corporation Farbbildverarbeitungsgerät
EP0270087A2 (de) * 1986-12-03 1988-06-08 Konica Corporation Farbbildverarbeitungsgerät

Family Cites Families (18)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US2598504A (en) * 1948-04-23 1952-05-27 Times Facsimile Corp Color picture transmission and reproduction
US3995312A (en) * 1975-02-28 1976-11-30 Seiscom Delta Inc. Color dot display
DE2853511C2 (de) * 1978-12-12 1981-11-19 Dr.-Ing. Rudolf Hell Gmbh, 2300 Kiel Verfahren und Vorrichtung zur Herstellung von Farbauszügen insbesondere für den Textildruck
US4415925A (en) * 1981-02-05 1983-11-15 Canon Kabushiki Kaisha Color original readout apparatus
JPS57208422A (en) * 1981-06-18 1982-12-21 Fuji Photo Film Co Ltd Hue judging device
JPS5817777A (ja) * 1981-07-24 1983-02-02 Fuji Xerox Co Ltd カラー画像読取装置
JPS58142675A (ja) * 1982-02-18 1983-08-24 Sanyo Electric Co Ltd カラ−画像処理方式
JPS58150371A (ja) * 1982-03-02 1983-09-07 Sony Corp カラ−プリンタにおける着色特性補正回路
JPS58153454A (ja) * 1982-03-08 1983-09-12 Fuji Xerox Co Ltd 多機能複写機
JPS58157255A (ja) * 1982-03-13 1983-09-19 Dainippon Screen Mfg Co Ltd 画像走査記録方法
JPS59128872A (ja) * 1983-01-14 1984-07-25 Fuji Xerox Co Ltd カラ−画情報処理装置
JPS59147572A (ja) * 1983-02-10 1984-08-23 Fuji Xerox Co Ltd カラ−画情報処理装置
JPS59163980A (ja) * 1983-03-08 1984-09-17 Canon Inc カラ−画像処理装置
JPS6063197A (ja) * 1983-09-17 1985-04-11 Ricoh Co Ltd 多色印字方法
JPS60148280A (ja) * 1984-01-12 1985-08-05 Fuji Xerox Co Ltd カラ−原稿読み取り装置
JPS60180378A (ja) * 1984-02-28 1985-09-14 Fuji Xerox Co Ltd 多色原稿読み取り装置
JPS61161091A (ja) * 1985-01-08 1986-07-21 Fuji Photo Film Co Ltd 画像処理方法
US4841361A (en) * 1987-12-03 1989-06-20 Konica Corporation Color image processing apparatus

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPS56166666A (en) * 1980-05-28 1981-12-21 Ricoh Co Ltd Color picture processor
JPS60160774A (ja) * 1984-01-31 1985-08-22 Fuji Xerox Co Ltd ２色原稿読み取り装置
EP0270090A2 (de) * 1986-12-03 1988-06-08 Konica Corporation Farbbildverarbeitungsgerät
EP0270087A2 (de) * 1986-12-03 1988-06-08 Konica Corporation Farbbildverarbeitungsgerät

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
PATENT ABSTRACTS OF JAPAN, vol. 6, no. 52 (E-100)[930], 7th April 1982; & JP-A-56 166 666 (RICOH K.K.) 21-12-1981 *
PATENT ABSTRACTS OF JAPAN, vol. 9, no. 382 (E-369), 24th December 1985; & JP-A-60 160 774 (FUJI XEROX K.K.) 22-08-1985 *

Cited By (28)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4907078A (en) * 1985-09-11 1990-03-06 Ricoh Company, Ltd. Method for reproducing color images which are distinguishable in a document having only two colors
EP0270090A2 (de) * 1986-12-03 1988-06-08 Konica Corporation Farbbildverarbeitungsgerät
EP0270090A3 (en) * 1986-12-03 1990-02-28 Konica Corporation Color image processing apparatus
US5016096A (en) * 1987-09-21 1991-05-14 Konica Corporation Apparatus for color processing specifically inside or outside a detected region
EP0765075A3 (de) * 1988-06-20 1997-06-04 Canon Kk Bildwiedergabe- und Bildverarbeitungsgerät
EP0350919A2 (de) * 1988-07-15 1990-01-17 Konica Corporation Gerät zur Farbbildherstellung
EP0350930A2 (de) * 1988-07-15 1990-01-17 Konica Corporation Farbbildverarbeitungsgerät
EP0350919A3 (en) * 1988-07-15 1990-09-26 Konica Corporation Color image processing apparatus
US5130791A (en) * 1988-07-15 1992-07-14 Konica Corporation Color image processing apparatus
EP0350930A3 (de) * 1988-07-15 1991-05-08 Konica Corporation Farbbildverarbeitungsgerät
EP0357407A2 (de) * 1988-08-30 1990-03-07 Kabushiki Kaisha Toshiba Bildsignalverarbeitungsgerät zur Verwendung in einem Farbbildwiedergabesystem
EP0357407A3 (de) * 1988-08-30 1992-04-08 Kabushiki Kaisha Toshiba Bildsignalverarbeitungsgerät zur Verwendung in einem Farbbildwiedergabesystem
US5268771A (en) * 1989-02-15 1993-12-07 Hitachi, Ltd. Image restoration processing system and method
EP0388877A3 (de) * 1989-03-22 1992-09-02 Konica Corporation Farbbildverarbeitungsgerät mit Bilddiskriminierungsmitteln
EP0388877A2 (de) * 1989-03-22 1990-09-26 Konica Corporation Farbbildverarbeitungsgerät mit Bilddiskriminierungsmitteln
EP0395032A2 (de) * 1989-04-28 1990-10-31 Konica Corporation Farbbildverarbeitungsgerät
EP0395032A3 (de) * 1989-04-28 1991-04-03 Konica Corporation Farbbildverarbeitungsgerät
US5206719A (en) * 1989-05-10 1993-04-27 Canon Kabushiki Kaisha Image processing apparatus including means for extracting an outline
EP0396980A3 (de) * 1989-05-12 1991-08-07 Dainippon Screen Mfg. Co., Ltd. Verfahren zur elektronischen Bildverarbeitung von Linienumrandungen
EP0396980A2 (de) * 1989-05-12 1990-11-14 Dainippon Screen Mfg. Co., Ltd. Verfahren zur elektronischen Bildverarbeitung von Linienumrandungen
EP0422602A2 (de) * 1989-10-12 1991-04-17 Hitachi, Ltd. Verfahren und Vorrichtung zur Farbbildaufzeichnung
EP0422602A3 (en) * 1989-10-12 1992-05-13 Hitachi, Ltd. Method of and apparatus for recording color image
US5386223A (en) * 1989-10-12 1995-01-31 Hitachi, Ltd. Method of apparatus for recording color image by overlapping toner images of different colors
EP0458978A4 (en) * 1989-12-15 1992-07-22 Hitachi, Ltd. Picture filing device and method thereof
US5231482A (en) * 1989-12-15 1993-07-27 Hitachi, Ltd. Image filing apparatus and method for thereby encoding and storing various documents
EP0458978A1 (de) * 1989-12-15 1991-12-04 Hitachi, Ltd. Vorrichtung und verfahren zum aufbewahren von bildern
EP0484890A3 (en) * 1990-11-06 1993-05-05 Dainippon Screen Mfg. Co., Ltd. Method of and apparatus for producing overlapping image area
EP0484890A2 (de) * 1990-11-06 1992-05-13 Dainippon Screen Mfg. Co., Ltd. Verfahren und Vorrichtung zur Erzeugung eines überlappenden Bildgebietes

Also Published As

Publication number Publication date
EP0251278B1 (de) 1995-11-29
US4980758A (en) 1990-12-25
DE3751614D1 (de) 1996-01-11
EP0251278A3 (en) 1990-02-28
DE3751614T2 (de) 1996-03-28

Similar Documents

Publication Publication Date Title
EP0251278A2 (de) Bildverarbeitungsverfahren und Vorrichtung dazu
US5153925A (en) Image processing apparatus
US4893188A (en) Document image entry system
US7002709B1 (en) Image processing system, image processing method, and image input system
KR940005349B1 (ko) 화상가공장치
JPS6198069A (ja) 画像処理装置
US6014462A (en) Image processing apparatus and method for distinguishing alphanumeric symbols on a white background and those on a mesh pattern and individually processing such image data
EP0500174B1 (de) Verfahren zur Bildverarbeitung und Abtastsystem/Ausdrucksystem zur Durchführung dieses Verfahrens
US6775031B1 (en) Apparatus and method for processing images, image reading and image forming apparatuses equipped with the apparatus, and storage medium carrying programmed-data for processing images
US5267330A (en) Image processing apparatus
US4996602A (en) Image-processing apparatus
US5031035A (en) Color image processing apparatus with color ghost correction
US5422735A (en) Image processing method and image processing apparatus
US5760929A (en) Image processing apparatus for processing discriminated color regions within specified boundaries
US5301039A (en) Image processing apparatus with pixel tone discrimination
JP2000092327A (ja) 画像形成装置
US6134355A (en) Binarization using a local average, and using error diffusion
US6228125B1 (en) Image processing apparatus for parallel image processing and method therefor
US6078405A (en) Image processing device
JP3989636B2 (ja) 画像処理装置
JP3265100B2 (ja) ファクシミリ装置
JP4047356B2 (ja) 画像処理装置
US20020063710A1 (en) Image processing apparatus and image processing method
EP0878957B1 (de) Bildverarbeitungsgerät
JP3304463B2 (ja) 画像形成装置

Legal Events

Date Code Title Description
PUAI Public reference made under article 153(3) epc to a published international application that has entered the european phase

Free format text: ORIGINAL CODE: 0009012

17P Request for examination filed

Effective date: 19870724

AK Designated contracting states

Kind code of ref document: A2

Designated state(s): DE FR GB IT NL SE

PUAL Search report despatched

Free format text: ORIGINAL CODE: 0009013

AK Designated contracting states

Kind code of ref document: A3

Designated state(s): DE FR GB IT NL SE

RAP3 Party data changed (applicant data changed or rights of an application transferred)

Owner name: KONICA CORPORATION

17Q First examination report despatched

Effective date: 19920702

RBV Designated contracting states (corrected)

Designated state(s): DE GB

GRAA (expected) grant

Free format text: ORIGINAL CODE: 0009210

AK Designated contracting states

Kind code of ref document: B1

Designated state(s): DE GB

REF Corresponds to:

Ref document number: 3751614

Country of ref document: DE

Date of ref document: 19960111

PLBE No opposition filed within time limit

Free format text: ORIGINAL CODE: 0009261

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: NO OPPOSITION FILED WITHIN TIME LIMIT

26N No opposition filed
PGFP Annual fee paid to national office [announced via postgrant information from national office to epo]

Ref country code: GB

Payment date: 19980622

Year of fee payment: 12

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: GB

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 19990629

GBPC Gb: european patent ceased through non-payment of renewal fee

Effective date: 19990629

PGFP Annual fee paid to national office [announced via postgrant information from national office to epo]

Ref country code: DE

Payment date: 20000626

Year of fee payment: 14

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: DE

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20020403